Quantum Sensing
Quantum sensing has seen a strong technology push over recent decades, but the market pull has not kept pace.
Successful field validation still does not automatically convert into procurement. The gap between what quantum sensors can demonstrate and what buyers are ready to purchase varies so widely by technologies and use cases that no single statistic captures it.
A clear illustration of the commercialisation challenge is DARPA’s ROCkN optical clock. The clock has been tested on aircraft and ground vehicles and completed a 3-week naval deployment in the Pacific. However, as of March 2026, DARPA was still seeking sufficient customer demand to justify production at scale [2]. That demand signal had not yet been supplied by any acquisition programme in the Army, Navy or Air Force. DARPA then intervened in August 2026 to finance the next phase, giving IonQ a contract to set up pilot production capacity for the clock [28, 29]. The contract supports the establishment of pilot-scale manufacturing rather than consistent procurement demand, as there is no declared military requirement or programme of record for the clock.
The same pattern appears outside the defence industry. Atomionics, a Singapore-based start-up, field-tested its cold-atom gravimeter with Rio Tinto Exploration in February 2025 and later ran evaluations backed by BHP Ventures [20, 21]. In September 2025, Atomionics raised $12.7 million in Pre-Series A funding to move from industrial trials towards scalable production. However, that funding shows the transition from pilot to product and still does not prove constant customer demand.
Neither case implies that the technology failed. Instead, both point to a different question: what has to exist around a working sensor for a succesful field test to lead to repeat procurement?
The widely cited Technology Readiness Level (TRL) assesses whether the hardware can be built to specification and used outside the laboratory. It does not show whether an organisation can integrate, operate and maintain the resulting product.
To measure this readiness gap, we use the Application Readiness Level (ARL). The framework uses NASA’s 9-step ARL concept as a starting point [1], then translates it into 7 procurement conditions relevant to quantum sensing: integration, test method, identifiable buyer, contract route, trained operators, long-term support and operating procedures. The readiness scores are analyst assessments based on public evidence rather than official NASA or programme scores.
Can the hardware be built and run outside the laboratory?
Can a buyer specify, purchase and support it?
Using the two frameworks, we scored 13 publicly documented applications, and they show a consistent readiness gap: application readiness trails technical readiness in 11 of 13. The two exceptions are long-established products that reached operational maturity years ago and already sit within established procurement markets.
Airborne caesium magnetometers (E1) and rubidium/caesium timing systems for networks and data centres (E3) are both routinely deployed and procured, placing them at TRL 9 / ARL 9.
Wider gaps do not imply weaker sensor performance. More often, the hardware has advanced further than the buying system around it: the purchasing power, specifications and contracting routes that would turn a working sensor into a market are not yet in place. A cold-atom gravimeter may outperform a mature sensor on sensitivity while remaining harder to procure because the application has not yet been translated into a clear product requirement and qualification process.
More broadly, industry matters as much as technology. Both cases where TRL and ARL are closely aligned serve established commercial markets with broad customer bases, while every 2- or 3-level gap appears in defence or aerospace applications shaped by individual procurement programmes rather than repeat commercial demand. Magnetic navigation is the aerospace exception: it shows only a 1-level gap, because it can use the existing RTCA DO-160 civil certification standard [3, 7].
A sensor type describes what physical quantity is measured, the modality describes the underlying quantum effect and operating principle used to measure it. That distinction matters commercially because the same physical quantity can be measured using technologies with very different components, operating conditions, integration requirements, levels of maturity and underlying supply chains.
In this section, quantum magnetometry is used as an illustration to highlight the inherent complexity of the quantum sensing market: an optically pumped vapour-cell magnetometer, a SQUID and an NV-diamond device all measure magnetic fields, but they sit at different TRL and ARL levels. The same sensor type can thus span technologies ranging from decades-old airborne survey systems to portable devices still in operator trials, with very different supplier bases and qualification pathways.
For buyers, the comparison ultimately happens at the system level. Sensitivity alone does not determine whether a sensor is deployable. Buyers judge whether the performance advantage survives the practical constraints of operating the complete system. A component-level performance advantage can disappear once the supporting hardware and platform integration are included.
The modality choice also changes the route to adoption. It determines whether the buyer needs cryogenic cooling and whether the instrument can fit on a drone or requires a larger vehicle. The commercial comparison therefore has to move from sensor performance to system performance. The decision can be expressed conceptually as:
DARPA’s Robust Quantum Sensors programme addresses this system-level challenge directly. It focuses on the environmental robustness and platform integration issues that often limit field performance, including vibration, motion and electromagnetic interference, while requiring sensor developers to work with platform integrators early in the development cycle [24]. The objective is not simply to improve laboratory sensitivity, but to preserve usable sensor performance under operational conditions.
Quantum sensing companies have to solve these integration, qualification and industrialisation challenges with a comparatively small capital base. QED-C estimated total quantum industry revenue at $1.9 billion in 2025, including $1.4 billion from quantum computing and $470 million from quantum sensing [25, 26], and the funding imbalance is even larger: of the $12.6 billion invested in quantum start-ups in 2025, McKinsey reports that 90 percent went to quantum computing, leaving sensing and communications to share the remaining 10 percent [27]. Compared to quantum computing companies, quantum sensing companies have substantially less capital available to build everything required between a successful prototype and a scalable product, while operating in smaller, application-specific markets, where fragmented demand and varying integration and qualification requirements make commercialisation more complex. This fragmentation and complexity are well illustrated by the magnetometry example. There are at least eight quantum sensor types, each with different modalities, main applications, operating conditions, supply chains, and routes to commercial adoption.
Addressing this complexity requires supply and demand to develop together. Sensor companies lack clear demand signals, while buyers and funders lack comparable evidence of which technologies are genuinely ready. Both sides are making decisions with incomplete information. Better readiness evidence and reusable test results can reduce that information gap, making credible opportunities easier to identify and fund.
To make things even more complicated, every use case brings its own buyer, specification and platform environment. Readiness thus has to be assessed at the application level, not inferred from the sensing modality alone, adding another layer of complexity.
An optically pumped magnetometer at TRL 7 has an ARL of 7 in magnetic navigation for crewed aircraft due to existing recognised integration partners, a certification basis and a certified product [3, 7]. On the other hand, anti-submarine detection scores ARL 5 since testing is still under way and open sources show no established acquisition programme for a quantum detector [8].
The chart shows how the same field-deployable OPM technology can reach different levels of application readiness. Magnetic navigation progresses further because more of the commercialisation pathway is already in place: RTCA DO-160 provides an established basis for environmental qualification, flight trials have been completed, and integration partners are identifiable [3, 7]. Anti-submarine detection remains 2 ARL levels lower because field testing has not yet developed into a comparable qualification, integration and procurement pathway [8].
The key takeaway is that TRL travels with the technology, while ARL travels with the mission. Once the hardware becomes fieldable, further progress depends increasingly on whether the application has a workable integration, qualification and procurement pathway. Those conditions can mature at very different speeds even when the underlying sensor architecture is unchanged.
Anti-submarine magnetic anomaly detection (MAD) shows why that distinction matters. The US Navy began receiving the AN/ASQ-81 in 1970, and roughly 1,585 units of that sensor and its successor were produced through 2002 [4]. Decades of operational use have already built the operating and sustainment system around that sensor.
That is why a single TRL or modality-level maturity score can be misleading commercially. The same OPM can be close to operational adoption in one mission and several readiness steps behind in another. For market assessment, the relevant unit of analysis is therefore the sensor technology × application pair, not the sensor technology in isolation.
Exhibit 4 provides a snapshot of application readiness across industries and sensor technologies. Each cell shows the highest readiness level supported by publicly available evidence for a specific technology-industry combination. The purpose is not to assign a single readiness score to an industry, but to show where a visible route from pilot to deployment already exists and where it does not.
Readiness is concentrated rather than evenly distributed. Sixteen cells reach ARL 7 or above, but 6 of these sit in Academia. Of the 10 higher-readiness commercial cells, 8 are concentrated in vapour-cell clocks, optically pumped magnetometers and SQUIDs, which all are technologies with commercial activity that predates the current quantum-funding wave [22, 23]. Defence shows the broadest coverage across emerging technologies, while several other industries still have few publicly visible routes to procurement.
Existing buying systems appear to accelerate adoption. Cold-atom gravimeters, for example, reach ARL 7 in environmental monitoring and metrology because geodesy agencies and observatories already know how to specify absolute-gravity measurements, operate the equipment and fund replacement instruments [9]. The dashed cells represent the opposite condition: open sources do not yet identify a clear buyer, specification or procurement pathway around the technology.
Academia requires a separate interpretation. It shows a pathway for every technology because universities and research institutions can purchase individual instruments for experimentation. That demonstrates that a system can be bought and operated, but not necessarily that a scalable product market exists. A high academic ARL is therefore a weaker commercial signal than the same score in defence, aerospace or industrial markets.
Unlike a quantum computer, which can operate in a controlled environment, fieldability for quantum sensing is crucial since size, weight, power and cost matter as much as sensitivity. For instance, the U.S. Defense Innovation Unit’s (DIU) Farseer solicitation accepts a gravimeter 40 times less accurate in exchange for half the volume and lower power [16]. The field-proven comparison instrument is roughly 7 times heavier and draws 10 times the requested power [9]. Performance also degrades on a moving, vibrating platform, and without shared field tests, buyers cannot reliably compare these trade-offs.
These constraints are shaped by the components inside the sensor. SWaP-C follows from the lasers, vapour cells, magnetic shielding, vacuum and cooling hardware a sensor is assembled from, the so-called enabling technologies. Improving fieldability therefore requires more than improving the sensing element itself, it requires redesigning the wider system around it.
Quantum sensors are built using components such as narrow-linewidth lasers, vapour cells, magnetic shielding, cryogenics and quantum-grade diamond, often from a handful of eligible suppliers worldwide.
This creates a direct bottleneck for sensor production. Even if demand for a sensor increases, manufacturers may not be able to scale output because critical components cannot be supplied in sufficient volume. CNAS, for example, reports that one Minnesota company was once the only domestic commercial supplier of specialised chip-scale lasers used in compact quantum clocks and sensors [18].
The concern extends beyond individual companies. In 2025 we conducted a Reusable Quantum Enabling Supply Chain Tool (ReQuEST) analysis in the context of NATO’s Transatlantic Quantum Community. The study identified concentrated dependencies across the quantum enabling supply chain and recommended targeted measures to reduce these vulnerabilities, including investment in low-SWaP, environmentally robust components and systems, relevant fabrication infrastructure, and interoperable testbeds [36]. Stanford and the OECD have reached similar conclusions about dependence on specialised suppliers [34, 35]. These supply-chain challenges are increasingly addressed in institutional programmes, besides the aforementioned activities, the IEC/ISO JTC 3 has a dedicated quantum enabling technology working group, while DIU treats component technologies as a separate Farseer line of effort [16, 17]. All recognize that scaling quantum sensing requires investment across the enabling supply-chain.
A buyer needs a standard it can cite in a contract. IEC/ISO JTC 3, created in 2024, lists one published standard and ten in development [17]. However, we found no publicly available NATO STANAG or US military standard for qualifying a quantum sensor as such [19].
In practice, programmes often rely on existing platform standards. Aviation applications can use RTCA DO-160 for environmental qualification, while defence programmes commonly reference standards such as MIL-STD-461 for electromagnetic compatibility [3, 15]. These standards determine whether a system can operate safely and reliably on a platform, but they do not provide a common method for comparing the sensing performance of competing technologies. Buyers therefore still rely heavily on application-specific trials. Q-CTRL’s reported DO-160 qualification shows that existing standards can cover part of this pathway [3], suggesting that adapting established qualification frameworks may sometimes be more practical than creating entirely new quantum-specific standards.
Export regulation adds another layer. Many high-performance quantum sensors fall under dual-use controls. The top-end instrument in most rows is a dual-use controlled item: optically pumped magnetometers below 20 pT/√Hz, SQUID systems below 50 fT/√Hz, ground gravimeters better than 10 microgal, and space-qualified or high-stability atomic frequency standards [33]. Mature sensors have operated within these controls for decades, but for newer suppliers they can add licensing requirements and restrict the accessible international market.
The quantum sensing market cannot be supported through a single commercialisation strategy due to its inherent complexity. Each modality supports multiple sensor types and different use cases, all with distinct buyers, operating environments, technical requirements and procurement pathways. That makes it hard for national and international institutions, funding bodies and corporates to analyse the market as a whole. Buyers and funders should focus on specific applications and on relieving shared enabling bottlenecks, rather than treating quantum sensing as a single sector.
Quantum sensing is best understood as a portfolio of application markets rather than a single sector. The useful unit of analysis is the technology × application pair: what the sensor measures, which mission it serves, the platform conditions it must meet and the buying pathway around it. That view takes into account that a caesium magnetometer which has been a catalogue product for decades and an NV-diamond prototype in operator trials can sit under the same quantum-sensing label while being at very different stages of commercial readiness.
Since the path to market is application-specific, sensors move beyond pilots fastest where a functioning commercial and procurement system already surrounds the technology. Where that system is still forming, progress depends less on another isolated demonstration and more on building the conditions that allow later buyers to qualify, integrate and procure the product. Better transparency can help build that system. Clearer evidence of where each application stands makes it easier to direct procurement and investment towards the problems that are actually holding it back.
To support the quantum sensing market as a whole, investments should be made which focus on strengthening the supply chain of enabling technologies. In this area, priority should be given to environmentally robust and low-SWaP components since fieldability requirements substantially reduce the number of viable suppliers.
Quantum sensors are increasingly proving that they can work outside the laboratory, the next step is to make the surrounding market equally ready to specify, integrate, fund and buy them.
Working with us
This article identifies where quantum sensing commercial uptake is being held back. The next question is usually more specific: what does that mean for your technology, market or investment decision?
Public evidence can show where an application sits today. It cannot tell an organisation which market to prioritise, whether a supplier’s readiness claims hold up under scrutiny, or what has to change before a technology becomes commercially deployable.
That is where we work with clients directly. We support:
If this analysis has raised a question specific to your organisation, we can turn it into a focused follow-up project.
Start a Conversation